I think that the first two elements of the lawsuit, the request for "Declatory Judgement of Noninfringement of Copyrights" and for "Declatory Judgement of No Misappropriation of Trade Secrets" are going to go relatively fast.
In both requests, Red Hat uses a phrase "An actual controversy exists between SCO and Red Hat as to...". I believe that this actual controversy, if the court agrees, means that the declatory judgement request will probably be decided long before any actual court case. I think that it is a means of getting relatively quick relief from the wrongful actions of another.
If the court finds that there is no actual controversy between the two sides, then I think that the relief will likely have to wait for the case to go to trial.
My guess is that if the court finds for Red Hat in the declatory judgements, SCO is going to fold and the only thing left will be for them to try to settle the rest out of court for as little money as they can spend.
I don't know how long such a declatory finding should take, but my completely uneducated guess is that we could see it before the end of the year or early next year.
The other five counts are for torts (I think) that Red Hat (and many of the rest of us) allege that SCO has committed. These are false advertising, deceptive trade practices, unfair competition, tortious interference, and trade libel and disparagement.
If the court grants Red Hat the two declatory judgements, I can't see how SCO could hope to prevail on the remaining five counts.
They claim to have a "hundreds" of lines of code. That's half a days work, and certainly not worth three billion dollars. (Kernel programmers create 50 thousand lines of code per month).
50,000 lines of code per month? I really have trouble believing that.
A bit more than 10 years ago, I was a software developer at one company that employed quite a few excellent developers. One of the strangest things about that company was that the president of the company believed that counting lines of code was the software metric that mattered.
In spite of being the second oldest developer there (the oldest was two months older than me), most of the time I was either first or second in terms of lines of code developed over the previosu two weeks.
The best I ever had for a 2 week period was about 3,500 lines of code. Usually it was about 1,500 to 2,500 lines of code per 2 week period.
50,000 lines of code a month would be about 10 to 20 times my usual effort.
So I really don't believe 50,000 lines of code a month.
If they are working a 40 hour week, that would be over 1,000 lines of code an hour!
Seriously, because of spam issues, there have been many proposals for ways to replace SMTP or to modify it. Some of them are downright comical.
But it's going to take something a lot bigger than that to change anything.
Any replacement would have to be completely backwards compatible with SMTP for years to come. Many people would never switch. Others would switch only after seeing it in operation for a long time.
Since it would have to be completely backwards compatible for years, any spam getting through now would still get through for years to come.
I think that what might prod most to change would be if one of those crazy schemes of having the Post Office charge postage on e-mail were to be enacted. Then, you would see the creation of something else designed to skirt the definition of e-mail under the scheme.
What you can do is to come up with an optional private method that would not break SMTP. That way, those who didn't want to use it could get along just fine without it.
Such a scheme would probably work best if it were adopted by a large percentage of the Internet. For that reason, it should be usable for everything form small personal SMTP servers up to very large SMTP servers that handle millions of people.
For what it's worth, I've considered the idea of just rejecting all incoming e-mail to my accounts that is not encrypted.
I'd like to see something for ISPs where subscribers could subscribe to a music service if they wished, but the requests for downloads would be handled locally instead of over the Internet.
The savings in bandwidth costs for ISPs not located in areas where bandwidth is relatively cheap could easily pay for a machine with plenty of disk space to install the software and files on.
Around here, it's about $1,000 a month for a T-1 line. It wouldn't take that many users to use the service to free up one or more T-1 lines.
Step 1.5) After the allegedly copied code is revealed, a massive undertaking is done to determine the true source of the code and who owns the copyright over it. There were Caldera/SCO personnel contributing code to Linux. Could it be one of them? If so, I think they would have a decidedly tough time convincing anyone that there was a copyright violation even if it was their code.
There would also be the question of why did they continue to distribute the code under the GPL after they determined the code was copied from them?
And, of course, the big question over why did it take them so long to file a lawsuit?
So far, every indication is that the code in question was IBM's own code (and SGI's and maybe some others), not SCO's, and the entire thing is not a copyright issue, but a contract issue.
In that case, it is hard to even see that the code would have to be excised from Linux. It would not be that SCO had any rights to it except for a contract issue over whether they could forbid IBM from distributing it. But if they won, IBM would presumably be ordered to recompense them for their loss, not everyone running Linux.
Except, that is, if they have a contract with SCO that forbids them from releasing the code. That is one reason why I think SCO is trying to push everyone into purchasing a license. The other is to get money to survive through the lawsuit.
I believe that the judge assigned to the case (or has another been assigned in his place) actually found in favor of the defendant in a previous copyright lawsuit because the plaintiff made no effort to mitigate damages.
In particular, the defendant wrote a book incorporating part of the plaintiff's. He sent a copy of the book asking the plaintiff if it was okay with him.
The plaintiff made no effort to read the book at all and did not decide to file a lawsuit until, I think, the third book was published.
Someone else posted a link to a news story about the case a few weeks ago.
The odd thing is that in this case, copyrights are not even involved. Instead, it is a contract dispute between IBM and SCO in which SCO alleges that IBM released IBM's own code contrary to their contract.
In the unlikely event that SCO should prevail, their sole satisfaction will have to come from IBM. The individual and corporate users of Linux would not have caused SCO to suffer any damages at all. Barring any licenses with SCO, they would have no duty to SCO.
Maybe that's why SCO is trying to push licenses from corporate users. If the corporations sign restrictive licenses with SCO, then those licenses might remain in force even after the case is shown to be completely lacking in substance making those corporations who sign the licenses liable to SCO for all future use of Linux.
Maybe it's about SCO gaining market share through a very strange approach to marketing and not an attempt to get money from IBM after all.
I wonder if SCO could devise a license that forbids anyone who signs it to ever purchase any Unix or Linux from anyone else but SCO.
So if someone else's copyright or patent DOES get into Linux or OpenBSD, then they can sue YOU for using it w/out a license.
I'm curious if that is actually true.
What instances have there been where someone has purchased or otherwise legally acquired any kind of copyrighted work from someone with the authority to sell that work which was subsequently found to contain copyright violations and for which the purchasers had to destroy, return, quit using, or license the work?
I think that if you buy a copyrighted work where there is reason to be suspicious of it's legitimacy, you may be required to do something.
But not where you bought it from someone who had every apparent authority to distribute it.
At least one legal scholar, Eben Moglen, a professor of Law at Columbia University, is saying that copyright law applies to the act of copying a work, not to the use of the work:
Users don't need a license to use copyrighted programs anymore than they need to pay a copyright fee before reading Gone with the Wind.
If there is any such problem for the users, I think that it is going to have to be from somewhere besides Title 17.
It seems that SCO's FUD is not only designed to scare users into paying them for licenses they almost surely don't need, but to rewrite U.S. law as well.
I'm wondering when Gartner is going to start recommending everyone stay away from Windows because of InterTrust's patent infringement lawsuit against Microsoft.
That would seem to be a lot stronger case than SCO's unproven/unprovable/nonexistent/absurd claims against IBM.
Even if SCO's claims were on the level, wouldn't patent infringements be stronger than copyright infringements? Assuming SCO did have a case, why would they have a case against users of Linux who haven't violated SCO's copyrights (violations of copyright law are, I understand, issues of copying, something that most Linux users are not doing) and InterTrust not have a case against everyone with Microsoft software that is likely to be considered to be patent infringements of InterTrust's patents?
Is it because Gartner is succumbing to SCO's compaign of FUD? Or is Gartner in a position to lose money with the increased use of Linux? Or maybe just make more money if Microsoft dominates?
I wonder how frequently, if ever, a user who bought a legitimate copy of something that turned out to be an inadvertant copyright infringement has been held liable for that copy.
Don't assume that the problem you are having is typical with any version of BSD.
I use OpenBSD heavily and have used FreeBSD and NetBSD and I've never seen that type of performance problem on any of them. And my fastest BSD machine is only 233 MHz.
I did a quick test by creating a 23 MB file in one directory on a moderately loaded 233 MHz Compaq running OpenBSD 3.3 and I was able to copy it to another directory in 2.89 seconds.
The executives of a public company have certain fiduciary duties to their shareholders.
If the company is a heavy user of Linux, the costs of the SCO licenses could be extremely expensive.
If the executives of the company purchase the SCO license, they could possibly be opening themselves up to shareholder lawsuits accusing them of not performing their fiduciary duties to their stockholders.
For exmaple, if a company only has one or two Linux systems, the executives might find it worthwhile to go ahead and purchase the license. However, if the license fee is $700 per CPU and they have 5,000 CPUS, that would be $3,500,000 dollars in license fees -- easily enough to draw the wrath of their shareholders.
In other words, if a company has very many Linux systems, it might be a very good idea to discuss the issues with their lawyers before agreeing to pay SCO a penny.
There was also a recent discussion of this point elsewhere, but I don't remember where.
The gist of the discussion was that if you buy something from someone who has every apparent authority to sell it to you, it doesn't matter if they didn't, at least in copyright issues.
Professor Eben's argument is along those lines. He's saying that if you bought it in good faith from someone who you truly believed had the authority to sell it to you, then the copyright holder has no claim against you for possessing it.
I think that if you buy something from someone who does not have that apparent authority or whose authority is questionable, you would be liable.
I suspect that publishing an unsubstantiated claim in a newspaper would probably not place questions on the authority of Red Hat, SuSE, Mandrake,..., to provide you with a legal copy of the software, but I may be wrong.
As for the GPL question goes, I agree that their continued distribution of Linux under the GPL should be sufficient for a judge to find against SCO. But it's much better, I think, to have a convincing case using several different arguments, not just one.
(a) Making of Additional Copy or Adaptation by Owner of Copy.
Notwithstanding the provisions of section 106, it is not an infringement for the owner of a copy of a computer program to make or authorize the making of another copy or adaptation of that computer program provided:
(1) that such a new copy or adaptation is created as an essential step in the utilization of the computer program in conjunction with a machine and that it is used in no other manner, or
Since you have to load it into memory to execute it, it is "an essential step in the utilization of the computer program in conjuction with a machine".
The possession of a work that infringes on a copyright does not appear to be a violation of the copyright laws at all.
So, even in the unlikely chance that SCO was able to demonstrate a violation of their copyrights, their actions might be against distributors, and they would have no right of action against the rest of us.
If there was merit to their claims, then there would be some small possibility that downloading the distribution instead of buying it at the store might count as making a copy and would then be an infringement. And so would making multiple copies of the distribution. But since there is nothing I can find in Title 17 about limiting software to a single computer, I bet you could still install Linux on a large number of machines from a single distribution set and not infringe.
As Eben Moglen (a law professor at Columbia University and general counsel to FSF) is quoted as saying:
Users don't need a license to use copyrighted programs anymore than they need to pay a copyright fee before reading Gone with the Wind. If you copy, distribute, or modify copyrighted material, then you can be in copyright violation,
But even if SCO did prevail on copyright issues, they might still not be able to go after users who make copies, distribute, or modify the material since for all practical purposes, it appears that we have the authority to do so.
I think SCO is going to have to show serious evidence that there is a copyright violation to remove that apparent authority.
The biggest problem with the Kernel guys redoing it, is now they can be considered 'tainted'.
That might be true for trade secret issues, but I don't think so for copyright issues.
For example, imagine that you were a writer and was writing a book about bullfighting. Having read Hemingway's Death In The Afternoon wouldn't taint you. You can use the ideas in the book as much as you wish as long as you don't just copy the sentences.
Or imagine you are a songwriter. Having listened to a love song (even if it was downloaded via Kazaa) does not mean that you are tainted can cannot write another love song.
I believe that copyrights are about the complete work, not to it's content or structure alone.
In fiction, there are, however, instances where you have to tread lightly. If you were to write your own Parry Hotter novel without permission from Mrs. Rowling, you would be in trouble because it would be a derivative work as a result of building upon the characters she has developed.
But in writing software, the situation is different because the new code would not be a derivative of the old code using the definition of derivative contained in Title 17 of the U.S. Code:
A ''derivative work'' is a work based upon one or more preexisting works, such as a translation, musical arrangement, dramatization, fictionalization, motion picture version, sound recording, art reproduction, abridgment, condensation, or any other form in which a work may be recast, transformed, or adapted. A work consisting of editorial revisions, annotations, elaborations, or other modifications which, as a whole, represent an original work of authorship, is a ''derivative work''.
As long as the new code is actually new, I don't think that it would be a derivative work even if those doing the work were intimately familiar with the old code.
The bigger question is whether there is any infringing code to start with.
SCO can claim copyright, but whether any Linux code infringes on it will have to be determined, possibly in court.
One other point. Several people have suggested that we can now go to the Copyright office and examine the code to see what infringes. That may or may not be true.
I've been told that when filing for a copyright on code, you don't necessarily have to include the entire code, but only the first and last few pages. The purpose is to reduce the amount of paper the copyright office has to store.
I don't know if that's for each program or for everything combined. And I might misunderstand the issue entirely. But if it is true, don't expect to learn anything meaningful by examining the code that was filed.
About 30 years ago, I took part in a psychological experiment that had to do with ink blots.
There were 4 test subjects and the psychologist in the room. He'd show an ink blot to each test subject in turn and record the responses.
I was test subject #4.
On the first ink blot, the first three all said the same thing and I said something different.
The second ink blot went like the first.
I remember that on one ink blot, the guy next to me tried to argue with me into agreeing with him, but I didn't.
In fact, in the entire series of ink blots, the only time I agreed with anyone else was the one time he asked me first. Then everyone else agreed with me.
It turned out that there was only one true test subject, test subject #4. The rest were in cahoots with the psychologist.
The purpose of the experiment was to measure our socialness. The psychologist was rather upset with me because I was way off the curve and told me that I was the most anti-social person he had ever met.
That's something coming from a psychologist who worked at a state reformatory.
Anyway, back on topic, I tend to use passwords that are quite long usually by stringing unusual words together or by creating nonsensical sentences. In both cases, unusual spelling, punctuation, and capitalization are present.
Slashdot Mirror
Arrgghh!
I can't divide!
If the are working a 40 hour week, that would be over 250 lines of code an hour or over 4 lines of code a minute.
I think that the first two elements of the lawsuit, the request for "Declatory Judgement of Noninfringement of Copyrights" and for "Declatory Judgement of No Misappropriation of Trade Secrets" are going to go relatively fast.
...". I believe that this actual controversy, if the court agrees, means that the declatory judgement request will probably be decided long before any actual court case. I think that it is a means of getting relatively quick relief from the wrongful actions of another.
In both requests, Red Hat uses a phrase "An actual controversy exists between SCO and Red Hat as to
If the court finds that there is no actual controversy between the two sides, then I think that the relief will likely have to wait for the case to go to trial.
My guess is that if the court finds for Red Hat in the declatory judgements, SCO is going to fold and the only thing left will be for them to try to settle the rest out of court for as little money as they can spend.
I don't know how long such a declatory finding should take, but my completely uneducated guess is that we could see it before the end of the year or early next year.
The other five counts are for torts (I think) that Red Hat (and many of the rest of us) allege that SCO has committed. These are false advertising, deceptive trade practices, unfair competition, tortious interference, and trade libel and disparagement.
If the court grants Red Hat the two declatory judgements, I can't see how SCO could hope to prevail on the remaining five counts.
50,000 lines of code per month? I really have trouble believing that.
A bit more than 10 years ago, I was a software developer at one company that employed quite a few excellent developers. One of the strangest things about that company was that the president of the company believed that counting lines of code was the software metric that mattered.
In spite of being the second oldest developer there (the oldest was two months older than me), most of the time I was either first or second in terms of lines of code developed over the previosu two weeks.
The best I ever had for a 2 week period was about 3,500 lines of code. Usually it was about 1,500 to 2,500 lines of code per 2 week period.
50,000 lines of code a month would be about 10 to 20 times my usual effort.
So I really don't believe 50,000 lines of code a month.
If they are working a 40 hour week, that would be over 1,000 lines of code an hour!
Thanks for posting that.
Now I have something to read while I eat supper.
Have you ever tried to subscribe to a mailing list on an account that doesn't admit anything that is not white-listed?
For what it's worth, I use procmail to implement a white-list on an account that I use only for mailing list subscriptions.
It is nearly impossible to subscribe to a mailing list on that account without shutting down the white list for a while.
No more mailing lists?
All a spammer would have to do is insert a few lines of random garbage that is different for each message.
I thought SMTP was Spam Mail Transfer Protocol!
Just kidding.
Seriously, because of spam issues, there have been many proposals for ways to replace SMTP or to modify it. Some of them are downright comical.
But it's going to take something a lot bigger than that to change anything.
Any replacement would have to be completely backwards compatible with SMTP for years to come. Many people would never switch. Others would switch only after seeing it in operation for a long time.
Since it would have to be completely backwards compatible for years, any spam getting through now would still get through for years to come.
I think that what might prod most to change would be if one of those crazy schemes of having the Post Office charge postage on e-mail were to be enacted. Then, you would see the creation of something else designed to skirt the definition of e-mail under the scheme.
What you can do is to come up with an optional private method that would not break SMTP. That way, those who didn't want to use it could get along just fine without it.
Such a scheme would probably work best if it were adopted by a large percentage of the Internet. For that reason, it should be usable for everything form small personal SMTP servers up to very large SMTP servers that handle millions of people.
For what it's worth, I've considered the idea of just rejecting all incoming e-mail to my accounts that is not encrypted.
All we know is they filed a complaint with the U.S. District Court in Delaware.
What we don't know is the nature of the complaint.
I'd like to see something for ISPs where subscribers could subscribe to a music service if they wished, but the requests for downloads would be handled locally instead of over the Internet.
The savings in bandwidth costs for ISPs not located in areas where bandwidth is relatively cheap could easily pay for a machine with plenty of disk space to install the software and files on.
Around here, it's about $1,000 a month for a T-1 line. It wouldn't take that many users to use the service to free up one or more T-1 lines.
The derived work would be AIX, that is the System V code and IBM's modifications as a whole.
... would be unlikely to be considered to be derived works unless they actually contain System V code.
JFS, NUMA, RCU,
Ooooooops. I mispoke.
The original contract was between AT&T and IBM.
OSX is not Unix.
It may have been somewhat based on UNIX, but the way Apple mucked it up, it is decidedly not UNIX any more.
Anything requiring you to configure the OS through graphics interfaces instead of editing a file is just not UNIX.
One question I've wondered about is "Who interprets the contract?"
By that, I mean that the contract was not between IBM and SCO, it was assigned to SCO.
What if Novell announces that they interpreted it quite differently in a way that agrees more with IBM?
After all, the meeting of the minds was between IBM and Novell, not between IBM and SCO.
You didn't include all the steps.
Step 1.5) After the allegedly copied code is revealed, a massive undertaking is done to determine the true source of the code and who owns the copyright over it. There were Caldera/SCO personnel contributing code to Linux. Could it be one of them? If so, I think they would have a decidedly tough time convincing anyone that there was a copyright violation even if it was their code.
There would also be the question of why did they continue to distribute the code under the GPL after they determined the code was copied from them?
And, of course, the big question over why did it take them so long to file a lawsuit?
So far, every indication is that the code in question was IBM's own code (and SGI's and maybe some others), not SCO's, and the entire thing is not a copyright issue, but a contract issue.
In that case, it is hard to even see that the code would have to be excised from Linux. It would not be that SCO had any rights to it except for a contract issue over whether they could forbid IBM from distributing it. But if they won, IBM would presumably be ordered to recompense them for their loss, not everyone running Linux.
Except, that is, if they have a contract with SCO that forbids them from releasing the code. That is one reason why I think SCO is trying to push everyone into purchasing a license. The other is to get money to survive through the lawsuit.
I believe that the judge assigned to the case (or has another been assigned in his place) actually found in favor of the defendant in a previous copyright lawsuit because the plaintiff made no effort to mitigate damages.
In particular, the defendant wrote a book incorporating part of the plaintiff's. He sent a copy of the book asking the plaintiff if it was okay with him.
The plaintiff made no effort to read the book at all and did not decide to file a lawsuit until, I think, the third book was published.
Someone else posted a link to a news story about the case a few weeks ago.
Bravo!
You stated it very well.
The odd thing is that in this case, copyrights are not even involved. Instead, it is a contract dispute between IBM and SCO in which SCO alleges that IBM released IBM's own code contrary to their contract.
In the unlikely event that SCO should prevail, their sole satisfaction will have to come from IBM. The individual and corporate users of Linux would not have caused SCO to suffer any damages at all. Barring any licenses with SCO, they would have no duty to SCO.
Maybe that's why SCO is trying to push licenses from corporate users. If the corporations sign restrictive licenses with SCO, then those licenses might remain in force even after the case is shown to be completely lacking in substance making those corporations who sign the licenses liable to SCO for all future use of Linux.
Maybe it's about SCO gaining market share through a very strange approach to marketing and not an attempt to get money from IBM after all.
I wonder if SCO could devise a license that forbids anyone who signs it to ever purchase any Unix or Linux from anyone else but SCO.
I'm curious if that is actually true.
What instances have there been where someone has purchased or otherwise legally acquired any kind of copyrighted work from someone with the authority to sell that work which was subsequently found to contain copyright violations and for which the purchasers had to destroy, return, quit using, or license the work?
I think that if you buy a copyrighted work where there is reason to be suspicious of it's legitimacy, you may be required to do something.
But not where you bought it from someone who had every apparent authority to distribute it.
At least one legal scholar, Eben Moglen, a professor of Law at Columbia University, is saying that copyright law applies to the act of copying a work, not to the use of the work:
If there is any such problem for the users, I think that it is going to have to be from somewhere besides Title 17.
It seems that SCO's FUD is not only designed to scare users into paying them for licenses they almost surely don't need, but to rewrite U.S. law as well.
I'm wondering when Gartner is going to start recommending everyone stay away from Windows because of InterTrust's patent infringement lawsuit against Microsoft.
That would seem to be a lot stronger case than SCO's unproven/unprovable/nonexistent/absurd claims against IBM.
Even if SCO's claims were on the level, wouldn't patent infringements be stronger than copyright infringements? Assuming SCO did have a case, why would they have a case against users of Linux who haven't violated SCO's copyrights (violations of copyright law are, I understand, issues of copying, something that most Linux users are not doing) and InterTrust not have a case against everyone with Microsoft software that is likely to be considered to be patent infringements of InterTrust's patents?
Is it because Gartner is succumbing to SCO's compaign of FUD? Or is Gartner in a position to lose money with the increased use of Linux? Or maybe just make more money if Microsoft dominates?
I wonder how frequently, if ever, a user who bought a legitimate copy of something that turned out to be an inadvertant copyright infringement has been held liable for that copy.
Don't assume that the problem you are having is typical with any version of BSD.
I use OpenBSD heavily and have used FreeBSD and NetBSD and I've never seen that type of performance problem on any of them. And my fastest BSD machine is only 233 MHz.
I did a quick test by creating a 23 MB file in one directory on a moderately loaded 233 MHz Compaq running OpenBSD 3.3 and I was able to copy it to another directory in 2.89 seconds.
The executives of a public company have certain fiduciary duties to their shareholders.
If the company is a heavy user of Linux, the costs of the SCO licenses could be extremely expensive.
If the executives of the company purchase the SCO license, they could possibly be opening themselves up to shareholder lawsuits accusing them of not performing their fiduciary duties to their stockholders.
For exmaple, if a company only has one or two Linux systems, the executives might find it worthwhile to go ahead and purchase the license. However, if the license fee is $700 per CPU and they have 5,000 CPUS, that would be $3,500,000 dollars in license fees -- easily enough to draw the wrath of their shareholders.
In other words, if a company has very many Linux systems, it might be a very good idea to discuss the issues with their lawyers before agreeing to pay SCO a penny.
That was the opinion of a noted law professor.
..., to provide you with a legal copy of the software, but I may be wrong.
There was also a recent discussion of this point elsewhere, but I don't remember where.
The gist of the discussion was that if you buy something from someone who has every apparent authority to sell it to you, it doesn't matter if they didn't, at least in copyright issues.
Professor Eben's argument is along those lines. He's saying that if you bought it in good faith from someone who you truly believed had the authority to sell it to you, then the copyright holder has no claim against you for possessing it.
I think that if you buy something from someone who does not have that apparent authority or whose authority is questionable, you would be liable.
I suspect that publishing an unsubstantiated claim in a newspaper would probably not place questions on the authority of Red Hat, SuSE, Mandrake,
As for the GPL question goes, I agree that their continued distribution of Linux under the GPL should be sufficient for a judge to find against SCO. But it's much better, I think, to have a convincing case using several different arguments, not just one.
The law is rather clear on that issue, though.
From U.S. Code, Title 17, Chapter 1, Section 117:
Since you have to load it into memory to execute it, it is "an essential step in the utilization of the computer program in conjuction with a machine".
You are quite correct.
The possession of a work that infringes on a copyright does not appear to be a violation of the copyright laws at all.
So, even in the unlikely chance that SCO was able to demonstrate a violation of their copyrights, their actions might be against distributors, and they would have no right of action against the rest of us.
If there was merit to their claims, then there would be some small possibility that downloading the distribution instead of buying it at the store might count as making a copy and would then be an infringement. And so would making multiple copies of the distribution. But since there is nothing I can find in Title 17 about limiting software to a single computer, I bet you could still install Linux on a large number of machines from a single distribution set and not infringe.
As Eben Moglen (a law professor at Columbia University and general counsel to FSF) is quoted as saying:
But even if SCO did prevail on copyright issues, they might still not be able to go after users who make copies, distribute, or modify the material since for all practical purposes, it appears that we have the authority to do so.
I think SCO is going to have to show serious evidence that there is a copyright violation to remove that apparent authority.
That might be true for trade secret issues, but I don't think so for copyright issues.
For example, imagine that you were a writer and was writing a book about bullfighting. Having read Hemingway's Death In The Afternoon wouldn't taint you. You can use the ideas in the book as much as you wish as long as you don't just copy the sentences.
Or imagine you are a songwriter. Having listened to a love song (even if it was downloaded via Kazaa) does not mean that you are tainted can cannot write another love song.
I believe that copyrights are about the complete work, not to it's content or structure alone.
In fiction, there are, however, instances where you have to tread lightly. If you were to write your own Parry Hotter novel without permission from Mrs. Rowling, you would be in trouble because it would be a derivative work as a result of building upon the characters she has developed.
But in writing software, the situation is different because the new code would not be a derivative of the old code using the definition of derivative contained in Title 17 of the U.S. Code:
As long as the new code is actually new, I don't think that it would be a derivative work even if those doing the work were intimately familiar with the old code.
The bigger question is whether there is any infringing code to start with.
SCO can claim copyright, but whether any Linux code infringes on it will have to be determined, possibly in court.
One other point. Several people have suggested that we can now go to the Copyright office and examine the code to see what infringes. That may or may not be true.
I've been told that when filing for a copyright on code, you don't necessarily have to include the entire code, but only the first and last few pages. The purpose is to reduce the amount of paper the copyright office has to store.
I don't know if that's for each program or for everything combined. And I might misunderstand the issue entirely. But if it is true, don't expect to learn anything meaningful by examining the code that was filed.
About 30 years ago, I took part in a psychological experiment that had to do with ink blots.
There were 4 test subjects and the psychologist in the room. He'd show an ink blot to each test subject in turn and record the responses.
I was test subject #4.
On the first ink blot, the first three all said the same thing and I said something different.
The second ink blot went like the first.
I remember that on one ink blot, the guy next to me tried to argue with me into agreeing with him, but I didn't.
In fact, in the entire series of ink blots, the only time I agreed with anyone else was the one time he asked me first. Then everyone else agreed with me.
It turned out that there was only one true test subject, test subject #4. The rest were in cahoots with the psychologist.
The purpose of the experiment was to measure our socialness. The psychologist was rather upset with me because I was way off the curve and told me that I was the most anti-social person he had ever met.
That's something coming from a psychologist who worked at a state reformatory.
Anyway, back on topic, I tend to use passwords that are quite long usually by stringing unusual words together or by creating nonsensical sentences. In both cases, unusual spelling, punctuation, and capitalization are present.
20 characters just doesn't seem enough.